{"id":"W2987837502","doi":"10.48550/arxiv.1911.02897","title":"Efficacy of Pixel-Level OOD Detection for Semantic Segmentation","year":2019,"lang":"en","type":"preprint","venue":"arXiv (Cornell University)","topic":"Advanced Neural Network Applications","field":"Computer Science","cited_by":6,"is_retracted":false,"has_abstract":true,"ca_institutions":"University of Waterloo","funders":"","keywords":"Computer science; Artificial intelligence; Task (project management); Segmentation; Ranking (information retrieval); Pixel; Metric (unit); Image (mathematics); Pattern recognition (psychology); Image segmentation; Computer vision; Engineering","routes":{"ca_aff":true,"ca_fund":false,"ca_venue":false,"about_ca":false,"invisible_to_affiliation_only":false},"retraction":null,"screen":null,"direct_labels":[],"prediction":{"model_version":"codex-gemma-dda1882f352a","candidate_categories":[],"consensus_categories":[],"category_scores_codex":[0.0001031458,0.000198278,0.000242556,0.0001765967,0.0001030122,0.00002943548,0.0009467867,0.0001570416,0.000002985726],"category_scores_gemma":[0.00002437085,0.0002422768,0.0001764048,0.0004823851,0.00004732854,0.0002719056,0.0006363774,0.0002164562,0.00002932121],"about_ca_system_candidate":false,"about_ca_system_consensus":false,"about_ca_system_score_codex":0.0001396505,"about_ca_system_score_gemma":0.00007471332,"about_ca_topic_candidate":false,"about_ca_topic_consensus":false,"about_ca_topic_score_codex":0.00001548349,"about_ca_topic_score_gemma":0.00002179693,"domain_scores_codex":[0.9986717,0.00004929115,0.0002088957,0.0007860992,0.00007111236,0.0002128624],"domain_scores_gemma":[0.9981373,0.0002608046,0.000403836,0.0009662106,0.0001700759,0.00006179771],"domain_codex":null,"domain_gemma":null,"domain_candidate":null,"domain_consensus":null,"study_design_codex":"simulation_or_modeling","study_design_gemma":"simulation_or_modeling","study_design_scores_codex":[0.0000628991,0.0001212797,0.0002216738,0.0001682336,0.00006817958,0.000002484007,0.0001154979,0.9440189,0.006455339,0.04087522,0.0000484683,0.007841809],"study_design_scores_gemma":[0.001244488,0.0001185304,0.002897415,0.00008594595,0.0001046437,0.000002515029,0.00002764429,0.9395128,0.0148775,0.04048507,0.0002449923,0.0003984986],"study_design_candidate":"simulation_or_modeling","study_design_consensus":"simulation_or_modeling","genre_codex":"methods","genre_gemma":"empirical","genre_scores_codex":[0.1878369,0.00001705966,0.8106099,0.00003802437,0.0003393545,0.0009196574,0.00001729239,0.0001143311,0.0001074253],"genre_scores_gemma":[0.9830858,0.00005634678,0.01621239,0.00002828931,0.00005201407,0.000006767701,0.00002582446,0.00001629054,0.0005163371],"genre_candidate":"empirical","genre_consensus":null,"teacher_disagreement_score":0.7952489,"threshold_uncertainty_score":0.987976,"prediction_status":"machine_predicted_unvalidated"},"machine_scores":{"provisional":true,"baseline":true,"maturity_gate_passed":false,"score_opus":0.09812074925881781,"score_gpt":0.22490034947419,"score_spread":0.1267796002153722,"validation_status":"score_only:v0-immature-baseline","note":"Baseline scores from an immature model (maturity gate not passed). Scores rank; they never assert a category."}}